Summary. Our proposal addresses the problem of advanced, PIK3CA-mutant breast cancer. We propose a role for PI3K-inhibitors in this disease at a stage when the disease no longer responds to endocrine manipulation, cdk4/6- or mTOR-inhibition and when patients typically receive palliative chemotherapy. These pre-treated tumors frequently display increased glucose avidity as evidenced by high SUVmax values in 18FDG- PET-scans and highly active glycolysis, also known as the Warburg effect. 30 - 40% of breast tumors have activating PIK3CA mutations, and these co-occur with expression of estrogen receptor (ER). There are current clinical trial efforts aimed using modern PI3K-inhibitors (PI3Ki) GDC0032, Copanlisib or Gedatolisib to restore responsiveness to ER blockade, but studies to integrate these drugs at a later stage of PIK3CA-mutant BC are lacking. Here we propose a role for these more modern PI3K-inhibitors at later stages of the disease, when PIK3CA-mutant tumors have developed endocrine, cdk4/6i and mTORi resistance. We have recently discovered that PI3Ki work by disrupting an AKT-independent pathway that regulates glycolysis and sustains the Warburg effect, in addition to the disruption of canonical signaling pathways via AKT. In this pathway PI3K triggers the mobilization and release of aldolase from the cytoskeleton, and this in turn potentiates aldolase activity and the Warburg effect. This basic mechanism links PI3K signaling with cytoskeletal dynamics and enhanced glycolytic flux. To what extent PI3Ki efficacy is determined by the ability to inhibit aldolase and to suppress the Warburg effect in advanced cancer, is still unknown and will be examined here. We propose that the activating PIK3CA mutation fuels the carbohydrate metabolism of PIK3CA-mutant BC, even at the stage of endocrine resistance, hat the anti-metabolic activity of PI3K-inhibitors can be visualized using in vivo NMR and exploited to improve treatment outcomes in combination with anti- neoplastic agents. In the first aim we will introduce a mutant form of aldolase to determine the contribution of the PI3K-aldolase axis to PI3Ki-efficacy in PIK3CA-mutant breast cancer. In the second aim we will determine in vivo if the metabolic changes induced by PI3K-inhibition are predictive of cancer treatment responses. We will visualize lower glycolysis directly in tumors in vivo using NMR detection of 13C[1]-pyruvate metabolism and compare the predictive value of NMR with that of 18FDG-PET CT. Simultaneous studies of the flux of 13C-glucose and 13C[1]-pyruvate will be used to pinpoint mechanisms of resistance. In a third aim, we will develop treatment combinations that may improve the outcomes for women with metastatic BC. Given the vulnerability of BC cells in the early post-mitotic phase, we propose to explore PI3K-inhibition following infusion of taxanes or vinca alkaloids in mouse models of PIK3CA-mutant endocrine-resistant BC. Separately, we will explore if the concept of the PI3Ki/Parpi-combination that is currently in clinical trials in TNBC, can be promising in advanced PIK3CA-mutant BC.